Thermal power plant remote inspection early warning method and system under support of 5G communication

Abstract

The invention relates to the technical field of power plant early warning, and provides a thermal power plant remote inspection early warning method and system under the support of 5G communication. The method comprises the following steps: acquiring a monitoring module corresponding to target thermal power plant equipment and a remote inspection monitoring center; establishing a first 5G communication module and a second 5G communication module to respectively transmit inspection data with different emergency degrees; historical inspection data is collected, probability and criticality analysis is carried out on equipment abnormity, and an emergency degree index is generated; and sending the real-time inspection data to a monitoring center for abnormal early warning based on the emergency index, wherein the monitoring center comprises an identification model trained based on historical abnormal data. The technical problems that in the thermal power plant inspection process, due to the fact that traditional wired and wireless networks are insufficient in communication performance and equipment abnormity key recognition is not accurate, the inspection efficiency is low, and potential safety hazards cannot be handled in time are solved, and the technical effects that real-time remote inspection and abnormity early warning based on 5G communication are achieved, and the inspection efficiency and the early warning precision are improved are achieved.

Description

Technical Field

[0001] The present application relates to the field of power plant early warning technology, and specifically to a remote inspection and early warning method and system for thermal power plants supported by 5G communications. Background Art

[0002] With the deepening of the digital transformation of the thermal power industry, the inspection and early warning system of power plants faces the challenge of insufficient efficiency and safety. Traditional inspection methods mainly rely on manual regular inspections and limited monitoring equipment, which is difficult to meet the needs of real-time and accuracy. Due to the wide distribution of equipment, inspections require a lot of manpower and material resources, and the method of manually discovering anomalies is inefficient. In addition, the existing data transmission methods have significant limitations. The deployment of wired networks is inflexible, costly, and complex to maintain; wireless networks are limited by insufficient bandwidth, high latency, and poor stability, and cannot meet the requirements of thermal power plants for efficient and stable data transmission. At the same time, the current abnormal warning often lacks in-depth analysis of equipment abnormalities, and cannot accurately judge the probability of abnormal occurrence and key impacts, resulting in delayed warnings and affecting the safe operation of equipment. Summary of the invention

[0003] This application provides a remote inspection and early warning method and system for thermal power plants supported by 5G communications, aiming to solve the technical problems of low inspection efficiency and inability to promptly deal with safety hazards during the inspection of thermal power plants due to insufficient communication performance of traditional wired and wireless networks and inaccurate critical identification of equipment abnormalities.

[0004] In view of the above problems, this application provides a remote inspection and early warning method and system for thermal power plants supported by 5G communications.

[0005] The first aspect disclosed in the present application provides a remote inspection and early warning method for a thermal power plant supported by 5G communication, the method comprising: obtaining multiple monitoring modules corresponding to multiple power plant equipment in a target thermal power plant, and a corresponding remote inspection and monitoring center; establishing a first 5G communication module and a second 5G communication module for the multiple monitoring modules and the remote inspection and monitoring center, wherein the first 5G communication module is used to transmit inspection data corresponding to a first communication urgency identifier, and the second 5G communication module is used to transmit inspection data corresponding to a second communication urgency identifier; collecting multiple historical remote inspection data sets corresponding to the multiple monitoring modules, performing abnormal probability analysis on the multiple power plant equipment, and generating multiple abnormal probability indicators; performing abnormal criticality analysis on the multiple power plant equipment, and establishing multiple abnormal criticality indicators; performing communication urgency analysis on the multiple power plant equipment in combination with the multiple abnormal probability indicators and the multiple abnormal criticality indicators, and generating multiple communication urgency indicators; based on the multiple communication urgency indicators, sending multiple real-time inspection data corresponding to the multiple monitoring modules to the remote inspection and monitoring center for abnormal identification and early warning, wherein the remote inspection and monitoring center includes an abnormal identification model trained by historical abnormal data.

[0006] Another aspect disclosed in the present application provides a remote inspection and early warning system for thermal power plants supported by 5G communication, the system comprising: a remote monitoring acquisition unit: acquiring multiple monitoring modules corresponding to multiple power plant equipment in a target thermal power plant, and a corresponding remote inspection and monitoring center; a communication module establishment unit: establishing a first 5G communication module and a second 5G communication module for the multiple monitoring modules and the remote inspection and monitoring center, wherein the first 5G communication module is used to transmit inspection data corresponding to a first communication urgency mark, and the second 5G communication module is used to transmit inspection data corresponding to a second communication urgency mark; an abnormal probability analysis unit: collecting multiple historical remote inspection data corresponding to the multiple monitoring modules According to the data set, the multiple power plant equipments are subjected to abnormal probability analysis to generate multiple abnormal probability indicators; the abnormal criticality analysis unit is used to perform abnormal criticality analysis on the multiple power plant equipments to establish multiple abnormal criticality indicators; the communication urgency analysis unit is used to perform communication urgency analysis on the multiple power plant equipments in combination with the multiple abnormal probability indicators and the multiple abnormal criticality indicators to generate multiple communication urgency indicators; the abnormality identification and early warning unit is used to send multiple real-time inspection data corresponding to the multiple monitoring modules to the remote inspection and monitoring center for abnormality identification and early warning based on the multiple communication urgency indicators, wherein the remote inspection and monitoring center includes an abnormality identification model trained by historical abnormal data.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned remote inspection and early warning method for thermal power plants supported by 5G communication, the method first equips each device with a corresponding monitoring module and connects it to the remote inspection and monitoring center. In terms of communication modules, two different types of communication modules are constructed through the 5G network. The first communication module is used to transmit emergency inspection data with higher priority, and the second communication module processes relatively minor inspection data to ensure the efficiency and accuracy of data transmission; then, the historical inspection data generated by all monitoring modules are collected, the abnormal probability analysis of each device is performed, the possibility of abnormality is calculated, and relevant indicators are generated. At the same time, by evaluating the critical impact of equipment abnormalities on the overall operation of the power plant, abnormal criticality indicators are established; then, the communication urgency of equipment is comprehensively analyzed in combination with the probability of abnormality and its criticality, and the communication urgency index of each device is obtained. Based on these urgency indicators, the real-time inspection data of high-urgency equipment is preferentially transmitted to the monitoring center, and the abnormal recognition model trained by historical data in the monitoring center is used to analyze and warn the transmitted data, thereby realizing timely detection and prevention of equipment abnormalities and improving the intelligence and safety of thermal power plant inspection management.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0011] Figure 1 It is a flow chart of a remote inspection and early warning method for a thermal power plant supported by 5G communication in an embodiment.

[0012] Figure 2 This is an architecture diagram of a remote inspection and early warning system for a thermal power plant supported by 5G communication in one embodiment.

[0013] Explanation of the accompanying drawings: remote monitoring acquisition unit 1, communication module establishment unit 2, abnormality probability analysis unit 3, abnormality criticality analysis unit 4, communication urgency analysis unit 5, abnormality identification and early warning unit 6. DETAILED DESCRIPTION

[0014] The embodiments of the present application provide a remote inspection and early warning method and system for thermal power plants supported by 5G communications, so as to solve the technical problems of low inspection efficiency and failure to deal with safety hazards in a timely manner during the inspection of thermal power plants due to insufficient communication performance of traditional wired and wireless networks and inaccurate critical identification of equipment abnormalities.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1, as Figure 1 As shown, the present application provides a remote inspection and early warning method for a thermal power plant supported by 5G communication, the method comprising:

[0018] Obtain multiple monitoring modules corresponding to multiple power plant equipment in the target thermal power plant, as well as the corresponding remote inspection and monitoring center.

[0019] In an embodiment of the present application, in the target thermal power plant, the system terminal is equipped with corresponding monitoring modules for multiple key equipment (such as main boilers, superheaters, main steam turbines, speed governors, main generators, absorption towers, etc.). These modules are responsible for real-time monitoring of the operating status and related data of the equipment. At the same time, a remote inspection and monitoring center is established to centrally receive and process data from each monitoring module. A data communication link is formed between the monitoring module and the monitoring center to ensure real-time monitoring and management of the equipment status, thereby providing support for subsequent data analysis and abnormal warnings.

[0020] A first 5G communication module and a second 5G communication module are established for the multiple monitoring modules and the remote inspection and monitoring center, wherein the first 5G communication module is used to transmit inspection data corresponding to a first communication urgency identifier, and the second 5G communication module is used to transmit inspection data corresponding to a second communication urgency identifier.

[0021] In one embodiment, in order to achieve efficient transmission of monitoring data, two 5G communication modules with different functions are constructed between multiple monitoring modules and the remote inspection and monitoring center, namely the first 5G communication module and the second 5G communication module. The first communication module has a higher configuration and is used to give priority to the transmission of inspection data with higher urgency, such as abnormal boiler pressure, generator temperature exceeding the limit, and other data that may cause equipment failure or safety accidents. The second communication module is suitable for transmitting inspection data with lower urgency, such as regular records of routine operating parameters or equipment status. This layered design ensures the priority transmission of key data while optimizing the utilization efficiency of network resources.

[0022] Furthermore, the present application provides a method for establishing a first 5G communication module and a second 5G communication module for the multiple monitoring modules and the remote inspection and monitoring center, including:

[0023] Acquire a real-time 5G communication network between the multiple monitoring modules and the remote inspection and monitoring center; slice the real-time 5G communication network through 5G slicing technology, and establish the first 5G communication module and the second 5G communication module, whose communication reliability is higher than the preset reliability, wherein the first 5G communication module is a high-speed data transmission slice, and the second 5G communication module is a low-power wide area network slice.

[0024] Preferably, according to the actual equipment distribution and communication needs of the thermal power plant, 5G base stations and access equipment are deployed in the plant area to cover all monitoring modules and the remote inspection and monitoring center, and a stable real-time 5G communication network is established. Then, each monitoring module is connected to the 5G network through wireless access, and at the same time, the connection between the remote inspection and monitoring center and the 5G core network is ensured to complete the unified configuration of the network nodes; subsequently, according to the data transmission needs of the thermal power plant inspection, the data is divided into two types: high urgency and low urgency. The data with high urgency requires low-latency and high-bandwidth transmission capabilities, while the data with low urgency pays more attention to low power consumption and wide coverage characteristics. In the real-time 5G communication network, the slice management function is used to divide two virtual network slices, namely the first 5G communication module and the second 5G communication module. Among them, the first 5G communication module is configured as an enhanced mobile broadband (eMBB) slice, which provides high-bandwidth and low-latency transmission capabilities for high-urgency data, and the second 5G communication module The second 5G communication module is configured as a massive machine type computing (mMTC) slice, providing low power consumption and wide coverage capabilities for low-urgency data; then, the high-speed data transmission slice is bound to the emergency equipment monitoring module, such as the sensors of the boiler and steam turbine, and the low-power wide area network slice is bound to the environmental monitoring module and the conventional operation data acquisition equipment, ensuring that each monitoring module and its corresponding communication slice are connected through the 5G network, and the corresponding slice receiver is set in the remote inspection and monitoring center; then, the two slices are tested for performance respectively to verify whether their communication reliability is higher than the preset standard, including indicators such as latency, bandwidth and power consumption, and the high-speed slice is tested for its low latency and high throughput performance, and the low-power slice is tested for its coverage and energy efficiency ratio. If the verification is passed, the two slices are used as the first 5G communication module and the second 5G communication module respectively. Otherwise, the slice resources are reallocated, for example, reducing the resource allocation of the low-power slice to enhance the performance of the high-speed slice. Through the above process, two communication modules based on 5G slicing technology are constructed to realize the classified priority transmission of inspection data, while ensuring the efficient use of network resources and the real-time monitoring needs of equipment.

[0025] A plurality of historical remote inspection data sets corresponding to the plurality of monitoring modules are collected, an abnormal probability analysis is performed on the plurality of power plant equipment, and a plurality of abnormal probability indicators are generated.

[0026] In one embodiment, the historical database of the monitoring module is parsed to collect historical inspection data related to the power plant equipment. These data include records of the equipment under different operating conditions, such as changes in key parameters such as temperature, pressure, and current. Based on these historical data, the system terminal analyzes the abnormal occurrence of each device, evaluates the possibility of abnormalities in the equipment under specific conditions, counts the frequency of abnormal events, establishes equipment abnormality frequency change curves, and generates corresponding abnormality probability indicators for each device based on these curves. These indicators quantify the risk of abnormalities in the equipment during future operation, and provide data support for subsequent priority judgments and early warnings.

[0027] Furthermore, the present application provides collecting multiple historical remote inspection data sets corresponding to the multiple monitoring modules, performing abnormal probability analysis on the multiple power plant equipment, and generating multiple abnormal probability indicators, including:

[0028] Based on the multiple historical remote inspection data sets, abnormal frequency analysis is performed on the multiple power plant equipment to establish multiple abnormal frequency change curves; multiple next abnormal prediction times of the multiple power plant equipment are predicted using the multiple abnormal frequency change curves; the time distance between the current moment and the multiple next abnormal prediction times is compared to generate the multiple abnormal probability indicators.

[0029] Preferably, after obtaining a plurality of historical remote inspection data sets, the frequency of abnormal occurrence of each power plant equipment within a preset time window (such as every hour or every day) is counted based on these data, that is, a plurality of groups of data are divided from the historical remote inspection data set of each power plant equipment, each group of data corresponds to a time window, the number of data points that are not within the normal operating range in each group of data is counted, and the ratio of the counted number to the number of data points in the time window is calculated to obtain the abnormal frequency of the time window; then, based on the plurality of time windows and the plurality of abnormal frequencies of each power plant equipment, the plurality of time windows and the plurality of abnormal frequencies are input into an exponential model, and the exponential fitting model is established based on exponential fitting, and is expressed as P(t)=P 0 ·e kt , where P(t) is the abnormal frequency at time t, P 0 is the initial abnormal frequency, which is a parameter obtained through fitting, and k is the growth or decay rate, which is also a parameter obtained through fitting. In the fitting process, the natural logarithm of both sides of the model formula is first taken to convert it into a linear form ln(P(t))=ln(P 0 )+kt, and then input multiple time windows and multiple abnormal frequencies into the transformed formula for multiple fittings to obtain the intercept ln(P 0) and slope k, and then the exponential model before the conversion is obtained through exponential recovery. Based on this exponential model, the system terminal establishes the abnormal frequency change curve of the power plant equipment, and repeats the above process to obtain multiple abnormal frequency change curves; then, based on these abnormal frequency change curves, combined with the abnormal frequency threshold (determined based on business needs and expert decisions), multiple time points in the future that are greater than or equal to the abnormal frequency threshold for the first time are determined from the multiple abnormal frequency change curves, and these time points are used as the next abnormal prediction time of the corresponding power plant equipment; then, for each power plant equipment, the difference between the current moment and the corresponding next abnormal prediction time is calculated as the time distance, and the abnormal probability index of each power plant equipment is determined using the functional relationship between the time distance and the abnormal probability index, providing reliable data support for remote inspection and early warning.

[0030] Furthermore, the present application provides a mechanism for comparing the time distance between the current moment and the multiple next abnormality prediction times, and generating the multiple abnormality probability indicators in a manner that the time distance is inversely proportional to the abnormality probability indicator.

[0031] Optionally, at the current moment, the time interval between the next predicted abnormal occurrence time of each power plant equipment and the current time is calculated and set as the time distance, and then the abnormal probability index of each power plant equipment is obtained based on the function mechanism that is inversely proportional to the time distance and the abnormal probability. Through this mechanism, the abnormal risk level of each power plant equipment can be dynamically quantified. The larger the abnormal probability index of the equipment, the closer the power plant equipment may be to the time point when the abnormality occurs, and it needs priority attention and processing. This mechanism provides a scientific risk assessment basis for the remote inspection and monitoring center, which helps to achieve accurate early warning and efficient resource allocation.

[0032] An abnormal criticality analysis is performed on the multiple power plant equipment, and multiple abnormal criticality indicators are established.

[0033] In one embodiment, for multiple power plant equipment, the system terminal performs a critical analysis of their abnormal conditions to evaluate the impact of each power plant equipment abnormality on the overall operation of the power plant. Specifically, data on abnormal events are extracted from the historical operation records of the power plant equipment, including the abnormality type, frequency of occurrence, duration, loss caused, and the number of other power plant equipment affected by the abnormality. Subsequently, based on these historical abnormality records, the possible impact range and diffusion effect of each power plant equipment abnormality, as well as the possible impact of each power plant equipment abnormality on the power plant's production operations are analyzed, thereby obtaining multiple abnormal diffusion indicators and multiple abnormal operation impact indicators. Afterwards, the abnormal diffusion indicator and abnormal operation impact indicator of each power plant equipment are weighted and summed to obtain the abnormal criticality indicator of each power plant equipment to quantify the importance of the power plant equipment abnormality. The higher the abnormal criticality indicator, the more significant the impact of the power plant equipment abnormality on the power plant, and the more priority attention is needed, thereby providing data support for power plant equipment priority sorting and abnormality handling, and helping to improve the operating safety and management efficiency of the power plant.

[0034] Furthermore, the present application provides for performing abnormal criticality analysis on the plurality of power plant equipment and establishing a plurality of abnormal criticality indicators, including:

[0035] Acquire historical abnormal record data of the multiple power plant equipment; perform abnormal diffusion analysis on the multiple power plant equipment based on the historical abnormal record data to generate multiple abnormal diffusion indicators; perform abnormal operation impact analysis on the multiple power plant equipment based on the historical abnormal record data to generate multiple abnormal operation impact indicators; fuse the multiple abnormal diffusion indicators and the multiple abnormal operation impact indicators to generate the multiple abnormal key indicators.

[0036] Preferably, from the monitoring module corresponding to each power plant equipment, the system terminal extracts the historical abnormal data of each power plant equipment and adds it to the historical abnormal record data, including the time, frequency, type, duration, loss caused, and the number of other power plant equipment affected by the abnormality; then, based on the historical abnormal record data, the abnormal diffusion analysis of the corresponding power plant equipment is performed to determine the diffusion domain of each power plant equipment, and then based on these diffusion domains, multiple abnormal diffusion indicators are determined. The larger the abnormal diffusion indicator, the more equipment is affected. At the same time, the abnormal operation impact analysis of the historical abnormal record data is performed to determine the power plant loss caused by each power plant equipment, and based on these power plant losses, multiple abnormal operation impact indicators are determined. The larger the abnormal operation impact indicator, the more loss is caused; then, the abnormal diffusion indicator and the abnormal operation impact indicator of each power plant equipment are merged by weighted summation to obtain abnormal key indicators, providing reliable data support for power plant equipment management and abnormal handling.

[0037] Furthermore, the present application provides an abnormal diffusion analysis of the plurality of power plant equipment based on the historical abnormal record data to generate a plurality of abnormal diffusion indicators, including:

[0038] Extract first historical abnormal data corresponding to the first device in the historical abnormal record data; perform abnormal diffusion analysis on the first device based on the first historical abnormal data to generate a first diffusion domain, and so on to generate multiple diffusion domains corresponding to the multiple power plant equipment; calculate the ratio of the range of the multiple diffusion domains to the sum of the ranges of the multiple diffusion domains to generate the multiple abnormal diffusion indicators.

[0039] Optionally, the abnormal data related to the first device is extracted from the historical abnormal record data to form the first historical abnormal data, wherein the first device refers to any one of a plurality of power plant devices; then, the number of other power plant devices affected by each abnormal event is extracted from the first historical abnormal data of the first device, and these numbers of power plant devices are defined as the first diffusion domain of the first device, thereby completing the abnormal diffusion analysis of the first device; thereafter, according to the analysis method of the first device, the abnormal diffusion analysis is performed on each power plant device in turn, the historical abnormal data of each device is extracted, the number of devices affected under abnormal conditions is identified, and the corresponding diffusion domain is generated; then, the sum of the diffusion domains of all devices is calculated, and for each power plant device, the proportion of its diffusion domain size to the total diffusion domain is calculated to obtain the abnormal diffusion index of the power plant device. The larger the abnormal diffusion index, the wider the impact of the power plant device on other devices when it is abnormal, and the greater its harmfulness.

[0040] Furthermore, the present application provides an abnormal operation impact analysis of the plurality of power plant equipment based on the historical abnormal record data, and generates a plurality of abnormal operation impact indicators, including:

[0041] Based on the first historical abnormal data, the loss of the first power plant caused by the abnormality of the first equipment is analyzed, and so on, the multiple power plant losses corresponding to the multiple power plant equipment are continued to be obtained; the ratio of the multiple power plant losses to the sum of the multiple power plant losses is calculated to generate the multiple abnormal operation impact indicators.

[0042] Optionally, data related to losses when an abnormality occurs in the first device is extracted from the first historical abnormal data, including direct losses (such as equipment maintenance costs, parts replacement costs, etc.) and indirect losses (such as power generation losses, breach of contract compensation, etc.), and the first power plant loss of the first equipment is obtained by adding these losses; then, the above steps are repeated for other equipment in the power plant (second equipment, third equipment, etc.) in turn to calculate the losses caused to the power plant when each power plant equipment is abnormal; thereafter, the loss values ​​of all power plant equipment are added to calculate the total power plant losses, and then the power plant losses of each power plant equipment are compared with the total power plant losses to generate an abnormal operation impact index for each power plant equipment. The larger the abnormal operation impact index, the more significant the impact of the equipment abnormality on the overall operation of the power plant and the greater its harmfulness.

[0043] The communication urgency analysis of the multiple power plant equipment is performed in combination with the multiple abnormal probability indicators and the multiple abnormal criticality indicators to generate multiple communication urgency indicators.

[0044] In one embodiment, after obtaining multiple abnormal probability indicators and multiple abnormal critical indicators, the system terminal assigns weights to the abnormal probability indicators and abnormal critical indicators according to actual needs and expert decisions. For example, if the urgency assessment focuses more on the overall impact of the abnormality on the power plant, a higher weight can be assigned to the abnormal critical indicator. If more attention is paid to the probability of abnormality, a higher weight is assigned to the abnormal probability indicator. For each power plant equipment, the system terminal will perform communication urgency analysis based on the abnormal probability indicator and multiple abnormal critical indicators of the power plant equipment, calculate the communication urgency index of the power plant equipment by weighted summation, and sort the communication urgency indicators from high to low to determine the equipment that needs to transmit data first. Through this analysis process, the communication urgency index of the power plant equipment can comprehensively reflect the possibility of abnormality and the degree of impact of the abnormality on the power plant, help the system terminal dynamically adjust the data transmission priority, and ensure that the status information of the key equipment can be timely transmitted to the remote inspection and monitoring center for early warning processing.

[0045] Furthermore, the present application provides that after generating multiple communication urgency indicators, the method further includes:

[0046] Based on the multiple historical remote inspection data sets, abnormal parallel correlation analysis is performed on the multiple power plant equipment, and several parallel associated equipment combinations with abnormal parallel correlation greater than a preset correlation are established; for the several parallel associated equipment combinations, one associated equipment in the combination is randomly selected, and communication urgency index degradation optimization is performed on the remaining associated equipment in the combination to generate several index degradation optimization results; based on the several index degradation optimization results, the multiple communication urgency indices are optimized.

[0047] Preferably, the system terminal extracts the number of abnormal events of each power plant equipment from the historical remote inspection data set. When performing abnormal parallel correlation analysis on any two power plant equipment, the abnormal correlation of the two equipment is obtained by calculating the ratio of the number of simultaneous abnormal events of the two power plant equipment to the total number of abnormal events of the two equipment. Subsequently, all equipment pairs whose abnormal correlation is greater than a preset correlation threshold are extracted, and these equipment pairs are correlated and combined according to their direct connections to obtain a number of parallel correlated equipment combinations. For example, if the equipment pair R 1,2 , R 1,4 , R 3,5 , R 3,6 , R 4,9 If the abnormal correlation meets the requirements of the preset correlation threshold, then device 1, device 2, device 4, and device 9 form a parallel associated device combination, and device 3, device 5, and device 6 form a parallel associated device combination; then, in each parallel associated device combination, a power plant device is randomly selected as the key device, and its communication urgency index is retained, and then the communication urgency index of other power plant devices in the combination is downgraded to obtain several index downgrade optimization results. The degree of degradation is proportional to the correlation. The higher the correlation, the greater the degradation range, ensuring that communication resources are used for key devices first; then, the communication urgency index in the index downgrade optimization results of all parallel associated device combinations is used to replace the communication urgency index of the corresponding power plant equipment, and multiple optimized communication urgency indexes are generated. Through downgrade optimization, the load of the first communication module can be reduced, while ensuring the accuracy of inspection and avoiding the omission of abnormal data due to downgrade.

[0048] Furthermore, the present application also includes:

[0049] The plurality of parallel associated device combinations are stored in the remote inspection and monitoring center. When the remote inspection and monitoring center issues an abnormal warning to any device, a linkage warning is simultaneously issued to other devices in the parallel associated device combination of any device.

[0050] Optionally, the identified multiple parallel associated device combinations are stored in the database of the remote inspection and monitoring center. Each device combination records the specific association relationship and number of the devices therein. When the remote inspection and monitoring center detects that any device issues an abnormal warning, it will automatically query the stored associated device combination to find all other devices associated with the device. For the associated devices, a linkage warning will be triggered based on the strength of their association with the abnormal device. This linkage warning not only reminds related equipment that may be affected by the abnormality, but also provides inspection personnel with a more comprehensive reference for abnormal situations. Through the linkage warning, the monitoring center can capture possible chain abnormalities in advance, optimize the timeliness and accuracy of equipment failure handling, and ensure the safety and stability of power plant operation.

[0051] Based on the multiple communication urgency indicators, the multiple real-time inspection data corresponding to the multiple monitoring modules are sent to the remote inspection monitoring center for abnormality identification and early warning, wherein the remote inspection monitoring center includes an abnormality identification model trained by historical abnormality data.

[0052] In one embodiment, according to the generated multiple communication urgency indicators, the monitoring modules of the power plant are prioritized, and the monitoring modules that need to be processed first and their corresponding real-time inspection data are determined. The system terminal transmits the real-time inspection data of each monitoring module to the remote inspection and monitoring center through the 5G communication network in turn according to the communication urgency indicator. In the remote inspection and monitoring center, an intelligent anomaly recognition model trained based on historical abnormal data is deployed. The anomaly recognition model can quickly analyze the received real-time inspection data, identify abnormal data in the equipment operation status, and generate abnormal warning information based on these abnormal data. The warning information includes the abnormal equipment number, abnormal type, etc., so that maintenance personnel can take timely measures, thereby improving the inspection efficiency and the safety of power plant operation.

[0053] In the abnormality recognition model, the system terminal will collect historical abnormal data as label data, including abnormal parameters, etc. At the same time, it also collects corresponding historical inspection data as input data; then, by dividing the input data and label data, the training set and the verification set can be obtained, and the constructed initial abnormality recognition model is trained by the training set. After the training, the verification set is used to verify and evaluate the performance of the model; among them, the initial abnormality recognition model can be constructed based on support vector machine (SVM), long short-term memory network (LSTM), random forest, etc. Taking support vector machine as an example, first, use support vector machine to build an initial abnormality recognition model structure, including input layer, support vector machine classifier, decision boundary, etc. The core task of SVM is to separate the samples of normal equipment and abnormal equipment by constructing an optimal hyperplane. In this way, it can be identified whether the equipment is abnormal. Subsequently, the support vector machine model is trained using the training set. The input layer receives the historical inspection data of the equipment. The SVM classifier maximizes the interval between normal and abnormal equipment by finding the best hyperplane. During the training process, the SVM adjusts the parameters of the classifier according to the label of the data until it can optimally distinguish normal data from abnormal data. After the training, the model performance is tested using the validation set to evaluate the model's accuracy, precision, recall rate and other indicators in the equipment anomaly classification task. If the performance of the model meets expectations and can effectively distinguish normal data from abnormal data, the current initial anomaly recognition model is output as the final anomaly recognition model. If the accuracy does not meet the target, the hyperparameters such as the kernel function and penalty parameter of the SVM are adjusted to further optimize the model and improve the classification effect.

[0054] In summary, the embodiments of the present application have at least the following technical effects:

[0055] The embodiment of the present application first collects the operating data of the thermal power plant equipment through multiple monitoring modules, and establishes two 5G communication modules with the remote inspection monitoring center, which are used to transmit inspection data of different urgency respectively; then, based on the historical inspection data set, the equipment is analyzed for abnormal probability, and multiple abnormal probability indicators are generated. At the same time, the critical indicators of the equipment are established through abnormal criticality analysis, and then the communication urgency analysis of the equipment is performed in combination with the abnormal probability and the criticality indicators to generate communication urgency indicators. The real-time inspection data is transmitted to the remote monitoring center through 5G communication for abnormality identification and early warning. In addition, by analyzing the historical abnormal record data, abnormal diffusion indicators and operation impact indicators are generated, and the abnormal criticality indicators are further optimized. Through parallel correlation analysis, the abnormal correlation between devices is identified, and the communication urgency of the associated devices is downgraded and optimized, thereby reducing the communication pressure of some devices; finally, the parallel correlation information between devices is stored in the monitoring center. When an abnormality occurs, other associated devices are linked to issue an early warning to ensure the safe and stable operation of the thermal power plant equipment. These technical effects jointly solve the technical problems in the inspection process of thermal power plants, such as low inspection efficiency and inability to deal with safety hazards in a timely manner due to insufficient communication performance of traditional wired and wireless networks and inaccurate identification of critical equipment anomalies. They realize real-time remote inspection and abnormality warning based on 5G communication, and improve inspection efficiency and warning accuracy.

[0056] Embodiment 2 is based on the same inventive concept as the remote inspection and early warning method for a thermal power plant supported by 5G communication in the aforementioned embodiment. Figure 2As shown, the present application provides a remote inspection and early warning system for thermal power plants supported by 5G communication, and the system includes: a remote monitoring acquisition unit 1: obtaining multiple monitoring modules corresponding to multiple power plant equipment in the target thermal power plant, and a corresponding remote inspection and monitoring center; a communication module establishment unit 2: establishing a first 5G communication module and a second 5G communication module for the multiple monitoring modules and the remote inspection and monitoring center, wherein the first 5G communication module is used to transmit the inspection data corresponding to the first communication urgency mark, and the second 5G communication module is used to transmit the inspection data corresponding to the second communication urgency mark; an abnormal probability analysis unit 3: collecting multiple historical remote inspection data sets corresponding to the multiple monitoring modules, An abnormal probability analysis is performed on the multiple power plant equipment to generate multiple abnormal probability indicators; an abnormal criticality analysis unit 4: an abnormal criticality analysis is performed on the multiple power plant equipment to establish multiple abnormal criticality indicators; a communication urgency analysis unit 5: a communication urgency analysis is performed on the multiple power plant equipment in combination with the multiple abnormal probability indicators and the multiple abnormal criticality indicators to generate multiple communication urgency indicators; an abnormality identification and early warning unit 6: based on the multiple communication urgency indicators, multiple real-time inspection data corresponding to the multiple monitoring modules are sent to the remote inspection and monitoring center for abnormality identification and early warning, wherein the remote inspection and monitoring center includes an abnormality identification model trained by historical abnormal data.

[0057] Furthermore, the communication module establishing unit 2 is also used to execute the following method:

[0058] Acquire a real-time 5G communication network between the multiple monitoring modules and the remote inspection and monitoring center; slice the real-time 5G communication network through 5G slicing technology, and establish the first 5G communication module and the second 5G communication module, whose communication reliability is higher than the preset reliability, wherein the first 5G communication module is a high-speed data transmission slice, and the second 5G communication module is a low-power wide area network slice.

[0059] Furthermore, the abnormal probability analysis unit 3 is also used to perform the following method:

[0060] Based on the multiple historical remote inspection data sets, abnormal frequency analysis is performed on the multiple power plant equipment to establish multiple abnormal frequency change curves; multiple next abnormal prediction times of the multiple power plant equipment are predicted using the multiple abnormal frequency change curves; the time distance between the current moment and the multiple next abnormal prediction times is compared to generate the multiple abnormal probability indicators.

[0061] Furthermore, the abnormal probability analysis unit 3 is also used to perform the following method:

[0062] The time distances between the current moment and the multiple next abnormality prediction times are compared, and the multiple abnormality probability indicators are generated by a mechanism in which the time distance is inversely proportional to the abnormality probability indicator.

[0063] Furthermore, the abnormal criticality analysis unit 4 is also used to perform the following method:

[0064] Acquire historical abnormal record data of the multiple power plant equipment; perform abnormal diffusion analysis on the multiple power plant equipment based on the historical abnormal record data to generate multiple abnormal diffusion indicators; perform abnormal operation impact analysis on the multiple power plant equipment based on the historical abnormal record data to generate multiple abnormal operation impact indicators; fuse the multiple abnormal diffusion indicators and the multiple abnormal operation impact indicators to generate the multiple abnormal key indicators.

[0065] Furthermore, the abnormal criticality analysis unit 4 is also used to perform the following method:

[0066] Extract first historical abnormal data corresponding to the first device in the historical abnormal record data; perform abnormal diffusion analysis on the first device based on the first historical abnormal data to generate a first diffusion domain, and so on to generate multiple diffusion domains corresponding to the multiple power plant equipment; calculate the ratio of the range of the multiple diffusion domains to the sum of the ranges of the multiple diffusion domains to generate the multiple abnormal diffusion indicators.

[0067] Furthermore, the abnormal criticality analysis unit 4 is also used to perform the following method:

[0068] Based on the first historical abnormal data, the loss of the first power plant caused by the abnormality of the first equipment is analyzed, and so on, the multiple power plant losses corresponding to the multiple power plant equipment are continued to be obtained; the ratio of the multiple power plant losses to the sum of the multiple power plant losses is calculated to generate the multiple abnormal operation impact indicators.

[0069] Furthermore, the communication urgency analysis unit 5 is also used to perform the following method:

[0070] Based on the multiple historical remote inspection data sets, abnormal parallel correlation analysis is performed on the multiple power plant equipment, and several parallel associated equipment combinations with abnormal parallel correlation greater than a preset correlation are established; for the several parallel associated equipment combinations, one associated equipment in the combination is randomly selected, and communication urgency index degradation optimization is performed on the remaining associated equipment in the combination to generate several index degradation optimization results; based on the several index degradation optimization results, the multiple communication urgency indices are optimized.

[0071] Furthermore, the communication urgency analysis unit 5 is also used to perform the following method:

[0072] The plurality of parallel associated device combinations are stored in the remote inspection and monitoring center. When the remote inspection and monitoring center issues an abnormal warning to any device, a linkage warning is simultaneously issued to other devices in the parallel associated device combination of any device.

[0073] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0074] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0075] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

The remote inspection and early warning method for thermal power plants supported by 1.5G communication is characterized in that: include: Obtain multiple monitoring modules corresponding to multiple power plant equipment in the target thermal power plant, as well as the corresponding remote inspection and monitoring center; Establishing a first 5G communication module and a second 5G communication module for the multiple monitoring modules and the remote inspection and monitoring center, wherein the first 5G communication module is used to transmit the inspection data corresponding to the first communication urgency identifier, and the second 5G communication module is used to transmit the inspection data corresponding to the second communication urgency identifier; Collecting multiple historical remote inspection data sets corresponding to the multiple monitoring modules, performing abnormal probability analysis on the multiple power plant equipment, and generating multiple abnormal probability indicators; Performing abnormal criticality analysis on the plurality of power plant equipment and establishing a plurality of abnormal criticality indicators; performing communication urgency analysis on the plurality of power plant equipment in combination with the plurality of abnormal probability indicators and the plurality of abnormal criticality indicators to generate a plurality of communication urgency indicators; Based on the multiple communication urgency indicators, the multiple real-time inspection data corresponding to the multiple monitoring modules are sent to the remote inspection monitoring center for abnormality identification and early warning, wherein the remote inspection monitoring center includes an abnormality identification model trained by historical abnormality data.

2. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 1 is characterized in that: Perform abnormal criticality analysis on the multiple power plant equipment and establish multiple abnormal criticality indicators, including: Acquire historical abnormal record data of the plurality of power plant equipment; Based on the historical abnormal record data, an abnormal diffusion analysis is performed on the plurality of power plant equipment to generate a plurality of abnormal diffusion indicators; Based on the historical abnormal record data, abnormal operation impact analysis is performed on the multiple power plant equipment to generate multiple abnormal operation impact indicators; The plurality of abnormal diffusion indicators and the plurality of abnormal operation impact indicators are integrated to generate the plurality of abnormal key indicators.

3. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 2 is characterized in that: Based on the historical abnormal record data, an abnormal diffusion analysis is performed on the multiple power plant equipment to generate multiple abnormal diffusion indicators, including: Extracting first historical abnormal data corresponding to the first device from the historical abnormal record data; Performing an abnormal diffusion analysis on the first device based on the first historical abnormal data to generate a first diffusion domain, and so on to generate multiple diffusion domains corresponding to the multiple power plant devices; The ratio of the ranges of the plurality of diffusion domains to the sum of the ranges of the plurality of diffusion domains is calculated to generate the plurality of abnormal diffusion indicators.

4. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 3 is characterized in that: Based on the historical abnormal record data, abnormal operation impact analysis is performed on the multiple power plant equipment to generate multiple abnormal operation impact indicators, including: Analyze the first power plant loss caused by the first device being abnormal based on the first historical abnormal data, and so on, continue to obtain multiple power plant losses corresponding to the multiple power plant devices; The ratio of the losses of the plurality of power plants to the sum of the losses of the plurality of power plants is calculated to generate the plurality of abnormal operation impact indicators.

5. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 1 is characterized in that: Collect multiple historical remote inspection data sets corresponding to the multiple monitoring modules, perform abnormal probability analysis on the multiple power plant equipment, and generate multiple abnormal probability indicators, including: Performing abnormal frequency analysis on the plurality of power plant equipment based on the plurality of historical remote inspection data sets, and establishing a plurality of abnormal frequency change curves; Predicting multiple next abnormal prediction times of the multiple power plant equipment using the multiple abnormal frequency change curves; The time distances between the current moment and the multiple next abnormality prediction times are compared to generate the multiple abnormality probability indicators.

6. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 5, characterized in that: The time distances between the current moment and the multiple next abnormality prediction times are compared, and the multiple abnormality probability indicators are generated by a mechanism in which the time distance is inversely proportional to the abnormality probability indicator.

7. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 1, characterized in that: Establishing a first 5G communication module and a second 5G communication module for the multiple monitoring modules and the remote inspection and monitoring center includes: Acquire a real-time 5G communication network between the plurality of monitoring modules and the remote inspection and monitoring center; The real-time 5G communication network is sliced ​​by 5G slicing technology to establish the first 5G communication module and the second 5G communication module, both of which have communication reliability higher than the preset reliability, wherein the first 5G communication module is a high-speed data transmission slice, and the second 5G communication module is a low-power wide area network slice.

8. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 1, after generating a plurality of communication urgency indicators, further comprising: Based on the multiple historical remote inspection data sets, abnormal parallel correlation analysis is performed on the multiple power plant equipment to establish a plurality of parallel correlation equipment combinations whose abnormal parallel correlation is greater than a preset correlation; Randomly selecting one associated device from the plurality of parallel associated device combinations, performing downgrade optimization of communication urgency indicators on the remaining associated devices in the combination, and generating a plurality of indicator downgrade optimization results; The plurality of communication urgency indicators are optimized based on the plurality of indicator downgrading optimization results.

9. The remote inspection and early warning method for a thermal power plant supported by 5G communication as claimed in claim 8, further comprising: The plurality of parallel associated device combinations are stored in the remote inspection and monitoring center. When the remote inspection and monitoring center issues an abnormal warning to any device, a linkage warning is simultaneously issued to other devices in the parallel associated device combination of any device. 10.5G communication-supported remote inspection and early warning system for thermal power plants, characterized by: The steps for implementing the remote inspection and early warning method for a thermal power plant supported by 5G communication as described in any one of claims 1 to 9 include: Remote monitoring acquisition unit: obtain multiple monitoring modules corresponding to multiple power plant equipment in the target thermal power plant, as well as the corresponding remote inspection monitoring center; A communication module establishing unit: establishing a first 5G communication module and a second 5G communication module for the multiple monitoring modules and the remote inspection and monitoring center, wherein the first 5G communication module is used to transmit the inspection data corresponding to the first communication urgency identifier, and the second 5G communication module is used to transmit the inspection data corresponding to the second communication urgency identifier; Abnormal probability analysis unit: collects multiple historical remote inspection data sets corresponding to the multiple monitoring modules, performs abnormal probability analysis on the multiple power plant equipment, and generates multiple abnormal probability indicators; Abnormal criticality analysis unit: performs abnormal criticality analysis on the plurality of power plant equipments and establishes a plurality of abnormal criticality indicators; A communication urgency analysis unit: performing communication urgency analysis on the plurality of power plant equipment in combination with the plurality of abnormal probability indicators and the plurality of abnormal criticality indicators to generate a plurality of communication urgency indicators; Abnormal identification and early warning unit: Based on the multiple communication urgency indicators, the multiple real-time inspection data corresponding to the multiple monitoring modules are sent to the remote inspection and monitoring center for abnormal identification and early warning, wherein the remote inspection and monitoring center includes an abnormal identification model trained by historical abnormal data.

11. The remote inspection and early warning system for thermal power plants supported by 5G communication as claimed in claim 10, characterized in that: Abnormal criticality analysis unit, including: A historical record acquisition unit is used to acquire historical abnormal record data of the plurality of power plant equipments; An abnormal diffusion analysis unit: performing abnormal diffusion analysis on the plurality of power plant equipment based on the historical abnormal record data to generate a plurality of abnormal diffusion indicators; An abnormal operation impact analysis unit: performs abnormal operation impact analysis on the plurality of power plant equipment based on the historical abnormal record data, and generates a plurality of abnormal operation impact indicators; An indicator fusion unit is configured to fuse the plurality of abnormal diffusion indicators and the plurality of abnormal operation impact indicators to generate the plurality of abnormal key indicators.

12. The remote inspection and early warning system for thermal power plants supported by 5G communication as claimed in claim 11, characterized in that: Abnormal diffusion analysis unit, including: A historical abnormal data extraction unit: extracting first historical abnormal data corresponding to the first device from the historical abnormal record data; A first abnormal diffusion analysis unit: performing an abnormal diffusion analysis on the first device based on the first historical abnormal data to generate a first diffusion domain, and so on to generate multiple diffusion domains corresponding to the multiple power plant devices; The abnormal diffusion index calculation unit calculates the ratio of the ranges of the plurality of diffusion domains to the sum of the ranges of the plurality of diffusion domains to generate the plurality of abnormal diffusion indexes.

13. The remote inspection and early warning system for thermal power plants supported by 5G communication as claimed in claim 12, characterized in that: Abnormal operation impact analysis unit, including: A power plant loss analysis unit: analyzing the first power plant loss caused by the first device abnormality based on the first historical abnormal data, and by analogy, continuing to obtain multiple power plant losses corresponding to the multiple power plant devices; The abnormal operation impact index analysis unit calculates the ratio of the losses of the plurality of power plants to the sum of the losses of the plurality of power plants, and generates the plurality of abnormal operation impact indexes.

Images